Abstract: This research presents a new approach for improving range image segmentation, based on fuzzy regularization of the detected edges. First, a degraded version of the segmentation is produced by a new region growing-based algorithm. Next, the resulting segmentation is refined by a robust fuzzy classification of the pixels on the resulting edges which correspond to borders of the extracted regions. Pixels on the boundary between tow adjacent regions are labeled taking into account the two regions as fuzzy sets in the fuzzy classification stage, using an improved version of the Fuzzy C-Mean (FCM) algorithm. The process is repeated for all region boundaries in the image. Extensive tests were performed with real images from the ABW database. The experimental results illustrate the good impact of the fuzzy regularization on the segmentation results. The high quality of the segmentation results compared with those of some well known segmentation methods show the good potential of our fuzzy approach for improving range image segmentation.